Biology subjects

Consens, M. E.

Publications and source records attributed to Consens, M. E..

4 recordsLinked to original sources

bioRxiv2026

Evolutionary conditioning enables guided generation of functionally diverse enhancers

Deep learning has been instrumental in our understanding of how enhancers encode regulatory information in their DNA sequence and has demonstrated preliminary success with enhancer design. However, the prevailing approach for enhancer design, cell type label conditioning, depends on labeled data from massively parallel reporter assays, which only exists for a handful of cell types. We propose EnhancAR, an autoregressive model trained on sets of unaligned homologous enhancer sequences to learn the function of the enhancer conserved over evolution and generate sequences that resemble real homologs. By training EnhancAR on 1.7 million human enhancer homolog sets spanning 1,888 cell types, EnhancAR generates enhancers for a variety of contexts without being conditioned on a cell type label. We computationally validate that when conditioned on a set of enhancer homologs, EnhancAR generates novel and diverse sequences that preserve the functional properties of the homologs. By prompting EnhancAR with homologs for existing cell type specific enhancers, we design enhancers with similar predicted cell type specificity. We further demonstrate that when trained on length sorted homologs, EnhancAR can design enhancers shorter than the conditioning homologs that preserve the predicted activity. In summary, we find that leveraging evolutionary information in enhancer homologs enables a more flexible and general paradigm for designing enhancers with specific functions.

bioengineering↗

bioRxiv2026

Predicting evolutionary rate as a pretraining task improves genome language model representations

Genome language models (gLM) have the potential to further understanding of regulatory genomics without requiring labeled data. Most gLMs are pretrained using sequence reconstruction tasks inspired by natural language processing, but recent studies have shown that these gLMs often fail to capture biological signal. To overcome this, we introduce pretraining tasks that predict the rate of evolution. These tasks are designed so that they can be composed with sequence reconstruction, enabling a controlled comparison of predicting sequence only, evolutionary rate only, or both. To address gaps in existing evaluations, we developed a suite of biologically grounded benchmarks. Across these tasks, and for established variant effect prediction benchmarks, models pretrained on both sequence and evolutionary rate outperform those trained on sequence alone, and training on evolutionary rate can make the even the relatively small models in our work competitive with much larger existing gLMs for some tasks. These results establish evolution as a key training target for genome-scale models.

genomics↗

bioRxiv2025

Interpreting Attention Mechanisms in Genomic Transformer Models: A Framework for Biological Insights

Transformer models have shown strong performance on biological sequence prediction tasks, but the interpretability of their internal mechanisms remains underexplored. Given their application in biomedical research, understanding the mechanisms behind these models predictions is crucial for their widespread adoption. We introduce a method to interpret attention heads in genomic transformers by correlating per-token attention scores with curated biological annotations, and we use GPT-4 to summarize each heads focus. Applying this to DNABERT, Nucleotide Transformer, and scGPT, we find that attention heads learn biologically meaningful associations during self-supervised pre-training and that these associations shift with fine-tuning. We show that interpretability varies with tokenization scheme, and that context-dependence plays a key role in head behaviour. Through ablation, we demonstrate that heads strongly associated with biological features are more important for task performance than uninformative heads in the same layers. In DNABERT trained for TATA promoter prediction, we observe heads with positive and negative associations reflecting positive and negative learning dynamics. Our results offer a framework to trace how biological features are learned from random initialization to pre-training to fine-tuning, enabling insight into how genomic foundation models represent nucleotides, genes, and cells.

bioinformatics↗

bioRxiv2022

Bulk and Single-nucleus Transcriptomics Highlight Intra-telencephalic and Somatostatin Neurons in Alzheimer's Disease

BackgroundCortical neuron loss is a pathological hallmark of late-onset Alzheimers disease (AD). However, it remains unclear which neuronal subtypes are most vulnerable to degeneration and contribute most to cognitive decline. MethodsWe analyzed postmortem bulk brain RNA-sequencing (RNAseq) data collected from three studies of aging and AD comprising six neocortical regions (704 individuals; 1037 samples). We estimated relative cell type proportions from each brain sample using neuronal subclass-specific marker genes derived from ultra-high depth single-nucleus RNAseq data (snRNAseq). We associated cell type proportions with AD across all samples using mixed-effects mega-analyses. Bulk tissue analyses were complemented by analyses of three AD snRNAseq datasets using the same cell type definitions and diagnostic criteria (51 individuals). Lastly, we identified cell subtype associations with specific neuropathologies, cognitive decline, and residual cognition. ResultsIn our mega-analyses, we identified the strongest associations of AD with fewer somatostatin (SST) inhibitory neurons ({beta}=-0.48, pbonf=8.98x10-9) and intra-telencephalic (IT) excitatory neurons ({beta}=-0.45, pbonf =4.32x10-7). snRNAseq-based cell type proportion analyses especially supported the association of SST neurons. Analyses of cell type proportions with specific AD-related phenotypes in ROS/MAP consistently implicated fewer SST neurons with greater brain-wide postmortem tau and beta amyloid ({beta}=-0.155, pFDR=3.1x10-4) deposition, as well as more severe cognitive decline prior to death ({beta}=0.309, pFDR=3.9x10-6). Greater IT neuron proportions were associated strongly with improved cognition ({beta}=0.173, pFDR=8.3x10-5) and residual cognition ({beta}=0.175, pFDR=1.2x10-5), but not canonical AD neuropathology. ConclusionsProportionally fewer SST and IT neurons were significantly associated with AD diagnosis across multiple studies and cortical regions. These findings support seminal work implicating somatostatin and pyramidal neurons in the pathogenesis of AD and improves our current understanding of neuronal vulnerability in AD.

neuroscience↗